In 2008, Bill Gates came to us with exactly such a problem. Despite breakthroughs in vaccine funding, research and manufacturing, these life-saving tools simply weren’t getting to enough of the children who needed them most. Today, the World Health Organization estimates that more than 1 million children under age five die each year from vaccine-preventable diseases—the majority of them in developing countries where it’s difficult to maintain the necessary stock of temperature-sensitive vaccines. There are a number of reasons why this is happening, but reliability of vaccine refrigeration and transport logistics are high on the list.
Never one to shy away from a challenge, our team at Intellectual Ventures set out to invent a better way. This four year journey took us from the confines of our laboratory to the stage of TED, the deserts of Afghanistan and rural Senegal, and now to the manufacturing floor of a leading refrigeration company. There is no victory until all the world’s children have access to life-saving vaccines, but the announcement today that AUCMA has agreed to manufacture and distribute our vaccine storage invention marks a significant milestone in the growth of IV Lab. We’ll dive deeper into the technical journey of the device in subsequent posts here on our blog, but I think it’s also important to acknowledge some of the guiding principles that helped us get here, what we learned along the way as a lab, and what we think we can do better next time.
Understand the Problem
We’re fortunate to have some incredible partners with deep expertise in developing world issues. What we learned—both through discussions here in Bellevue and in visits to developing countries—was that the core of this particular problem was deceptively difficult. How do we reach remote areas of developing countries with enough cold (0-8° C) vaccines to maintain regular immunization services? If the vaccines freeze or get too warm, they’re rendered ineffective, by WHO standards. What’s worse is that it isn’t always an easy thing to determine when vaccines have or haven’t spoiled, so it’s possible that many children are given vaccines without the required potency.
Before we started this project, we asked a simple question: money goes into a system, and some number of vaccinated children come out, but what happens within that system? We wanted to understand specifically what problems needed new ideas, and if our existing new ideas were likely to solve the problem. Unfortunately, there was a limited amount of information to guide us. There were thousands of people working on this high level problem across the world, but we discovered that things we take for granted here in Seattle, like the ability to measure costs or efficiencies within a system, are far more challenging in areas of the world plagued by real poverty, weak infrastructure, and where day-to-day survival is uncertain.
Similarly, in wealthier parts of the world, simple technologies are just part of our lives; it’s almost incomprehensible that it would be difficult to keep important things cold. We open the fridge on a hot day to get cold, unspoiled food, or we kick open the cooler at a camp site to get an ice-cold beverage. But anyone who has lost power at their house or ran out of ice while camping can relate to the difficulty of keeping vaccines cold in areas of the developing world without reliable electricity. In this case, though, shortcomings in the temperature-controlled vaccine supply chain can be the difference between life and death.
Understand What’s Already Been Done
The problem we hoped to solve may have been playing out half a world away, but the inspiration for our potential solution came from something close to home. At IV Lab, we often have cryogenic fluids (liquids that boil at less than 110 Kelvin) like liquid nitrogen on hand. These are stored in specially-designed containers called dewars, which have a double wall design with a high vacuum maintained between the layers to prevent heat transfer, to help their contents hold sub-zero temperatures. Two of our inventors, Drs Lowell Wood and Nathan Myhrvold, asked the question: if we can keep liquid nitrogen cold for weeks without power, what’s to stop us from using the same approach to store vaccines for a long time with only ice?
In fact, our first experiments were exactly that – we threw a bunch of ice in a cryogenic dewar to see how long it stayed frozen. As it turns out, not very long. The science that makes traditional dewars work at cryogenic temperatures doesn’t work so well with old fashioned ice. Nonetheless, the basic idea inspired us. If super-insulation techniques like vacuum-sealing, low conductivity materials selection, and multi layered insulation (MLI) could be used for a variety of commercial purposes, it seemed feasible that they could be adapted to strengthen and extend the vaccine cold chain in developing countries.
However, it became clear that we needed to deepen our expertise in the field of super-insulation. Our team spent months reading literature, calling and flying around the world to meet with cryogenic vessel experts, and sharing our concept ideas with healthcare workers and officials in remote areas of the world. While today we may be making headway with the release of our Passive Vaccine Storage Device, it’s clear that we built upon what others had done before us. We hope that others still will take what we’ve done and extend the technology, finding new ways to employ it in our world.
Iterate With Intent
The vaccine storage device went through six different prototype versions with dozens of iterations on each. Our approach to prototyping emphasizes the importance of using each iteration to address a specific question or problem. Early prototypes were focused almost exclusively on testing our hypothesis that super-insulation techniques could be applied to vaccine storage. This panned out for us, but not before countless modifications to test which combination worked best. There were also structural and material questions we had to answer. What should the device be made out of? Could it be made durable enough to withstand conditions in developing countries?
We made prototypes specifically so we could try to break them, which was equal parts entertaining and frustrating. It took a long time to fabricate each prototype – our machine shop team is one of the most capable I’ve ever seen, but we were learning how to make them in real time. In retrospect, we could have gone faster by increasing our ratio of intent:iterate.
What do I mean by that? One way to look at technology development is as a continuous effort to remove risk from an idea. We want to look at the idea from the vantage point of 5 years in the future and ask, “When this idea failed, what was its fatal flaw?” Then we want to make sure we don’t wait 5 years to figure it out! Even with the clarity of asking the hardest questions first, it’s not easy. Data may be king, but unfortunately, it’s not always easy to generate, or to understand it when we have it (alas, our experiments are designed by we knaves). Knowing what questions are important is critical, and knowing how to be critical of our questions is more so. On this project, we heavily utilized computer modeling tools to ask questions using electrons before cutting metal. Then, as each experiment was performed and data were collected, our model strengthened, leading to development of a simple and more easily used tool. We can now input desired key performance metrics, and it predicts combinations of design variables which will yield them.
Even now, we’re not done learning. We have half a dozen small projects underway, looking at ways to improve durability, reduce heat leak, and to combine our PVSD with other technologies which may enable different solutions in the field. The size and number of our technical risks are going down, however, allowing us to focus on reducing real-world risks.
Understand the Use Case
As we progressed through the prototyping process, we went from devices designed specifically to test technical concepts to those that explored functionality features like how health workers retrieved vaccines, changed ice or monitored temperature. For any invention, success in this stage hinges on a deep understanding not just of the technical challenge your invention needs to address, but also of the “who” and “how” of its use. For example, these factors influenced our decisions to move away from a vending-machine-style dispensing system and to sacrifice some hold time in favor of ease-of-access and durability.
This kind of trade is one of the hardest challenges for a technical team. Earlier embodiments of the technology allowed substantially less heat to leak in, which might have made the devices smaller and lighter than our eventual device; however, they lacked important user considerations which separate the “technically awesome” that engineers might drive towards from the “useful” that enables a real world impact. At the same time, users may not always have an understanding of what is possible. Demonstrations of capabilities other than those initially requested may unlock potential for true disruption, and far more meaningful impact. Iterations ensue.
In hindsight, this is another area where we could have done better earlier on in the process. Our partner organization on this project here at IV – Global Good – did not exist until we had been working on the cold chain device for two and a half years, which meant we initially didn’t have the level of on-the-ground issue and market expertise we now enjoy. Neither did we have as steady a focus on driving towards hand-off to an external partner, such as AUCMA, who can transform the idea to reality. As Global Good grew, so too did the pace of our progress through the product development phases. Today, we are working with the Global Good team to quickly iterate the learning cycles of what can be done and what should be done, developing use-case targets to shoot for from the outset of our projects. The Global Good team is also instrumental in managing the field testing and partnerships needed to translate breakthroughs in IV Lab to impact in developing countries.
Invention is a Collaborative Process
There’s no doubt that breakthrough “Eureka!” moments do occur for inventors, but the reality is that most inventions evolve from a much longer collaborative process. That was definitely the case for this device. What started as an idea by one or two of IV’s inventors was brought to life by more than 30 people who directly contributed to the project over the years. Each one brought something new to the effort. More than anything on this project, I’m proud of how we came together as a team, adapted to new challenges, and produced a device that has a chance to make a tangible impact on the lives of children in developing countries. Our device may only be one piece of the broader vaccine cold chain puzzle, but it goes to the developing world with all of the ingenuity, passion and hard work our team at IV Lab has to offer.